Brian Aker gave this 10 minute lightning talk on NoSQL at the Nov 2009 OpenSQLCamp in Portland, Oregon. It's incredibly funny, probably because there's a lot of truth to what he's saying.
Here are the slides and here are the notes. Found though #nosql.
This was really funny. I'm quite new to this fantastic website, but I've got a question : I'm not familiar with this "no SQL" school, but how does it relates to object oriented databases ?
I've always thought that the issue has always been having two different data stuctures : an object-oriented one for your business logic, and another one for your data structure. The language used to fetch data from a store isn't really the main problem, so I guess (but i can only try to guess) that no-sql really means no relationnal datastore, not "no store structure at all"... aka : persisting objects tree by simply saying "Save To Store" to its root object rather than doing multiple "insert / select identity" for as many tables you had to split your data into.
PS : i understand this video was meant to be funny before all, but it may be a good starting point for a real discussion.
Ugh. Distateful for those that are curious, but I'm sure self-satisfying for those in the club.
This is badly, deeply sick.
It's funny, but the jokes are all predicated on a straw man argument; it's not map-reduce all the way down.
By the way, how many visitors here do know how to write a distributed map reduce Erlang-function? And how many people decided to go fuck themselves because they can't?
A poll would be nice!
I've been laboring for an analogy for how this makes me feel. The best I've come up with is that I feel like I'm watching a bunch of dinosaurs giggle about those silly warm-blooded, hairy mammals. "Ho Ho Ho! Don't they realize that regulating your own body temperature is a waste of time?"
Wow, what a laugh!
Don't take Brian too seriously! He, like anyone else, sees the value of Map-reduce.
It is classic satire, which means it has just enough truth in it to make it funny.
And like any stand up routine, it is only funny because the presenter makes you believe that he really believes what he is saying... :)
> I'm watching a bunch of dinosaurs giggle about those silly warm-blooded, hairy mammals.Indeed, I felt the same too
Paul, I have my suspicions about Brian's motives here. He's clearly misrepresenting NoSQL and MapReduce, and as Samuel points out, he's attacking a straw man rather than understanding what the new paradigms have to offer. Granted, it might not have been funny if he presented it fairly, but it's not really funny anyway. It just looks like "giggling dinosaurs".
What are you talking about a bunch of Dinosaurs laughing at the warm blooded animals. Do you know how tried and true RDBMS are? For the last two decades every vendor out there has tried to dethrone the RDBMS with no success. Don't use the extremes like a google or amazon where they have hundreds of millions of dollars to design their own architectures either. Research network relational models or even odbms and you will see both were new and exciting and most of them all fell flat on their faces. I would be very surprised to see just how many of these new storage technologies come from developers who just get frustrated that they have to change their mindset to relations instead of objects in the database world.
Notify me of follow-up comments via email.